Modified Multi-Objective Particle Swarm Optimization: Application to Optimization of Diesel Engine Control Parameter
Dongmei Wu, Masatoshi Ogawa, Yasumasa Suzuki, Harutoshi Ogai, Jin Kusaka · SICE Journal of Control Measurement and System Integration · 2010
In this paper, optimization of diesel engine control parameters using a modified multi-objective particle swarm optimization (MOPSO) method is considered. This problem is formulized as a multi-objective optimization problem involving three optimization objectives: brake specific fuel consumption (BSFC), exhaust gas emission, and soot. A modified MOPSO is proposed with integration of particle swarm optimization (PSO) and a crossover approach. Several benchmark functions are tested, and results reveal that the modified MOPSO is more efficient than the typical MOPSO. Engine control parameter optimization with an extended PSO and the modified MOPSO is simulated, respectively. It proved the potential of the modified MOPSO for the engine control parameter optimization problem.